[PATCH v2 02/13] static_call: Add basic static call infrastructure

[Peter Zijlstra]

From: Josh Poimboeuf <jpoimboe@xxxxxxxxxx>

Static calls are a replacement for global function pointers.  They use
code patching to allow direct calls to be used instead of indirect
calls.  They give the flexibility of function pointers, but with
improved performance.  This is especially important for cases where
retpolines would otherwise be used, as retpolines can significantly
impact performance.

The concept and code are an extension of previous work done by Ard
Biesheuvel and Steven Rostedt:

  https://lkml.kernel.org/r/20181005081333.15018-1-ard.biesheuvel@xxxxxxxxxx, jpoimboe@xxxxxxxxxx
  https://lkml.kernel.org/r/20181006015110.653946300@xxxxxxxxxxx

There are two implementations, depending on arch support:

 1) out-of-line: patched trampolines (CONFIG_HAVE_STATIC_CALL)
 2) basic function pointers

For more details, see the comments in include/linux/static_call.h.

[peterz: simplified interface]
Signed-off-by: Josh Poimboeuf <jpoimboe@xxxxxxxxxx>
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
---
 arch/Kconfig                      |    3 
 include/linux/static_call.h       |  134 ++++++++++++++++++++++++++++++++++++++
 include/linux/static_call_types.h |   15 ++++
 3 files changed, 152 insertions(+)
 create mode 100644 include/linux/static_call.h
 create mode 100644 include/linux/static_call_types.h

--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -960,6 +960,9 @@ config RELR
 config ARCH_HAS_MEM_ENCRYPT
 	bool
 
+config HAVE_STATIC_CALL
+	bool
+
 source "kernel/gcov/Kconfig"
 
 source "scripts/gcc-plugins/Kconfig"
--- /dev/null
+++ b/include/linux/static_call.h
@@ -0,0 +1,134 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_STATIC_CALL_H
+#define _LINUX_STATIC_CALL_H
+
+/*
+ * Static call support
+ *
+ * Static calls use code patching to hard-code function pointers into direct
+ * branch instructions.  They give the flexibility of function pointers, but
+ * with improved performance.  This is especially important for cases where
+ * retpolines would otherwise be used, as retpolines can significantly impact
+ * performance.
+ *
+ *
+ * API overview:
+ *
+ *   DECLARE_STATIC_CALL(name, func);
+ *   DEFINE_STATIC_CALL(name, func);
+ *   static_call(name)(args...);
+ *   static_call_update(name, func);
+ *
+ * Usage example:
+ *
+ *   # Start with the following functions (with identical prototypes):
+ *   int func_a(int arg1, int arg2);
+ *   int func_b(int arg1, int arg2);
+ *
+ *   # Define a 'my_name' reference, associated with func_a() by default
+ *   DEFINE_STATIC_CALL(my_name, func_a);
+ *
+ *   # Call func_a()
+ *   static_call(my_name)(arg1, arg2);
+ *
+ *   # Update 'my_name' to point to func_b()
+ *   static_call_update(my_name, &func_b);
+ *
+ *   # Call func_b()
+ *   static_call(my_name)(arg1, arg2);
+ *
+ *
+ * Implementation details:
+ *
+ *    This requires some arch-specific code (CONFIG_HAVE_STATIC_CALL).
+ *    Otherwise basic indirect calls are used (with function pointers).
+ *
+ *    Each static_call() site calls into a trampoline associated with the name.
+ *    The trampoline has a direct branch to the default function.  Updates to a
+ *    name will modify the trampoline's branch destination.
+ *
+ *    If the arch has CONFIG_HAVE_STATIC_CALL_INLINE, then the call sites
+ *    themselves will be patched at runtime to call the functions directly,
+ *    rather than calling through the trampoline.  This requires objtool or a
+ *    compiler plugin to detect all the static_call() sites and annotate them
+ *    in the .static_call_sites section.
+ */
+
+#include <linux/types.h>
+#include <linux/cpu.h>
+#include <linux/static_call_types.h>
+
+#ifdef CONFIG_HAVE_STATIC_CALL
+#include <asm/static_call.h>
+/*
+ * Either @site or @tramp can be NULL.
+ */
+extern void arch_static_call_transform(void *site, void *tramp, void *func);
+#endif
+
+
+#define DECLARE_STATIC_CALL(name, func)					\
+	extern struct static_call_key STATIC_CALL_NAME(name);		\
+	extern typeof(func) STATIC_CALL_TRAMP(name)
+
+#define static_call_update(name, func)					\
+({									\
+	BUILD_BUG_ON(!__same_type(*(func), STATIC_CALL_TRAMP(name)));	\
+	__static_call_update(&STATIC_CALL_NAME(name),			\
+			     &STATIC_CALL_TRAMP(name), func);		\
+})
+
+#if defined(CONFIG_HAVE_STATIC_CALL)
+
+struct static_call_key {
+	void *func;
+};
+
+#define DEFINE_STATIC_CALL(name, _func)					\
+	DECLARE_STATIC_CALL(name, _func);				\
+	struct static_call_key STATIC_CALL_NAME(name) = {		\
+		.func = _func,						\
+	};								\
+	ARCH_DEFINE_STATIC_CALL_TRAMP(name, _func)
+
+#define static_call(name)	STATIC_CALL_TRAMP(name)
+
+static inline
+void __static_call_update(struct static_call_key *key, void *tramp, void *func)
+{
+	cpus_read_lock();
+	WRITE_ONCE(key->func, func);
+	arch_static_call_transform(NULL, tramp, func);
+	cpus_read_unlock();
+}
+
+#define EXPORT_STATIC_CALL(name)	EXPORT_SYMBOL(STATIC_CALL_TRAMP(name))
+#define EXPORT_STATIC_CALL_GPL(name)	EXPORT_SYMBOL_GPL(STATIC_CALL_TRAMP(name))
+
+#else /* Generic implementation */
+
+struct static_call_key {
+	void *func;
+};
+
+#define DEFINE_STATIC_CALL(name, _func)					\
+	DECLARE_STATIC_CALL(name, _func);				\
+	struct static_call_key STATIC_CALL_NAME(name) = {		\
+		.func = _func,						\
+	}
+
+#define static_call(name)						\
+	((typeof(STATIC_CALL_TRAMP(name))*)(STATIC_CALL_NAME(name).func))
+
+static inline
+void __static_call_update(struct static_call_key *key, void *tramp, void *func)
+{
+	WRITE_ONCE(key->func, func);
+}
+
+#define EXPORT_STATIC_CALL(name)	EXPORT_SYMBOL(STATIC_CALL_NAME(name))
+#define EXPORT_STATIC_CALL_GPL(key)	EXPORT_SYMBOL_GPL(STATIC_CALL_NAME(name))
+
+#endif /* CONFIG_HAVE_STATIC_CALL */
+
+#endif /* _LINUX_STATIC_CALL_H */
--- /dev/null
+++ b/include/linux/static_call_types.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _STATIC_CALL_TYPES_H
+#define _STATIC_CALL_TYPES_H
+
+#include <linux/stringify.h>
+
+#define STATIC_CALL_PREFIX	____static_call_
+#define STATIC_CALL_PREFIX_STR	__stringify(STATIC_CALL_PREFIX)
+
+#define STATIC_CALL_NAME(name)	__PASTE(STATIC_CALL_PREFIX, name)
+
+#define STATIC_CALL_TRAMP(name)	    STATIC_CALL_NAME(name##_tramp)
+#define STATIC_CALL_TRAMP_STR(name) __stringify(STATIC_CALL_TRAMP(name))
+
+#endif /* _STATIC_CALL_TYPES_H */